Analyze This: Boosting Communication Effectiveness With Contextual Segmentation

Segmentation is well understood as a concept but its potential is grossly underutilized, especially in pharma. The data and talent now exist to generate and execute on context-specific segments. Try it and you’ll see a significant boost in your marketing effectiveness.

Segmentation, in the context of marketing effectiveness, could be described as the classification of audiences into unique addressable groups to increase the communication relevance, impact, and customer value. Contrary to what you may have been told, segmentation does not have to be complex (though as an analytics advocate, I am aware that more variables could sometimes be better).

The underlying requirements of a good segmentation scheme are that group membership must be unique (i.e., have unique attributes common to the group and different from other groups), that members respond similarly to marketing stimulus, and that segments are targetable (no need segmenting on a variable you cannot identify and communicate on). As such, segmentation could be as simple as grouping patients on demographics, attributes, behavior, motivations, value, or as complex as combining these into multidimensional segmentation schemes.

While most pharma clients have adopted segmentation and some have proceeded even to create microsegments (subsegments) from these large segmentation schemes, many are yet to maximize the significant value of segmentation. The typical implication is that programs are created around segments and left to run. Tactics are developed and mapped back to these single segmentations and never reviewed or reassessed in light of real response and post-execution behavioral data. Rather than such single set-and-run or must-adhere-to-regardless-of-tactic segmentations, a more pragmatic segmentation should reflect the specific context, program, and feedback (or should we say shout-back) data that patients and HCPs are sending back. That’s what I mean by context-based segmentation.

Context-based segmentation could change and evolve depending on the communication, tactic, and platform. I reiterate that contextual segmentation is NOT microsegmentation, the subcategorization of larger segments. Contextual segmentation refers to creation of context-specific segments, which may be totally unhinged from any macro brand segment or its derivative.  Contextual segmentation takes on more importance especially in the absence of individually identifiable information. When you segment site visitors based on content selection, navigational paths, and visit frequency, you have just conducted some contextual segmentation. When you take this a step further and adapt content or offers to their needs, you are tapping that powerful potential of contextual segmentation. When you deliver offers to physicians based on historical response to multichannel communications, that’s segmentation within the context of a multichannel communication program. The result is the guarantee of communication effectiveness given the empirical feedback-based improvement in communication to your patient or physician.

So why has such a promising concept not been pervasive in marketing? It’s partly due to history and inertia. Until recently, two important components of executing contextual segments—data access and analytics expertise—were either totally limited or unavailable. But a lot has changed over the years: data has exploded—think Big Data—and many top-tier analytics and consulting practices like those at Ogilvy CommonHealth Worldwide (yes, shameless plug, I know…but it’s true) have superb analytics talents. But the industry has lagged in identifying and adapting to the reality.

It’s a fact that pharma data does have its unique challenges with data access: no tracking of consumer script behavior, few websites require log-ins, no shopping carts, unsure whether the intended doctor saw the sample shipment—the office staff may just have put your brochure out there in the waiting room. And let’s not forget that pharmaceutical firms are behind other verticals in social media activation. But you can make your own data as well. I don’t mean “make up” data (a no-no for any analytics person)—I mean you can create surveys, conduct interviews, observe patient-doctor relationships, purchase lists, and so on. Despite these limitations, you still have a rich base of datasets that can provide amazing segmentation schemes that can increase communication effectiveness 2-5x.

At OCHWW, we’ve generated several of these dramatic results first hand. Creating contextually based communication for a multichannel program based on patients’ engagements and channel preferences. We added transaction history with just a few survey questions administered by an IVR and created addressable segments for the brand. We then delivered separate communications streams designed specific to the right customer audience, and significantly raised campaign effectiveness.

I am not advocating the elimination of organizational or brand level segmentation. Those are valuable segments in themselves for specific purposes (Did I hear you say those are relevant for their respective context?). But pharmaceutical brands should embrace context-based segmentation. The excuse around lack of data or lack of talent is no longer valid. Forcing programs into organizational segments, or worse, not conducting any segmentation at all, should no longer be tolerated. You may have been unknowingly (or now that you know, knowingly) undermining the effectiveness of your program or campaign. On your next program, demand it.

A special thank you goes out to Leslie Prives for contributing to this blog post.

Questions? Comments? You can contact the author directly at
Please allow 24 hours for response.

This entry was posted in advertising, Analytics, Branding, Healthcare Communications, Marketing and tagged , , , , , , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.

One Comment

  1. Posted February 18, 2016 at 3:42 am | Permalink

    Spot on with this write-up, I absolutely think this website needs a lot more attention. I’ll probably be returning to see more, thanks for the information!